Skip to Main Content
Autonomous landing for Unmanned Aerial Vehicle is an important direction in the field of UAV research. Currently, relative GNC methods for aerial vehicle landing including GPS (Global Position System), INS (Inertial Navigation System) and ILS (instrument landing system) can't fully satisfy the requirements for UAV autonomous landing on the runway or un-cooperation environment. With the gradual progress and maturity of computer vision technology, visual navigation technology in UAV autonomous landing area has been extensively studied. In this paper, we firstly review different vision-based approaches for guidance and safe landing of different types of UAVs including fixed wing and rotorcraft and point out the key technology of vision-based GNC. Secondly, we introduce our research work in this field. In our group, aiming at the real airport scenes pictures taken from a UAV landing process, the real-time and robust method of image feature information extraction is proposed. As well, we realize the algorithm software in DM642 hardware platform and tested its practical application performance in the laboratory semi-physical simulation environment. The prospect for the development of vision-based GNC for UAV autonomous landing is made at the end of the paper.